TL;DR
This paper develops a mathematical model of the pituitary-thyroid feedback loop to investigate Allan-Herndon-Dudley Syndrome (AHDS). By incorporating membrane transporters, the model suggests that impaired thyroid hormone transport, leading to T4 retention in thyroid cells, can fully explain the characteristic hormonal imbalances seen in AHDS patients.
What Problem Does This Paper Address?
Allan-Herndon-Dudley Syndrome (AHDS) is a rare, severe genetic disorder characterized by distinct thyroid hormone levels: low free thyroxine (FT4), slightly elevated thyroid-stimulating hormone (TSH), and high free triiodothyronine (FT3). These abnormalities are linked to mutations in the SLC16A2 gene, which encodes the monocarboxylate transporter 8 (MCT8), crucial for thyroid hormone transport. A leading hypothesis suggests that the loss of MCT8 function causes thyroid hormone (specifically T4) to be retained within thyroid cells, disrupting the normal feedback mechanisms. However, the precise mechanisms and whether this retention alone can account for all observed hormonal changes remain areas of active investigation. This paper aims to address this by using a computational model to test this hypothesis.
Methods
The researchers extended an existing mathematical model of the pituitary-thyroid feedback loop. The core innovation is the inclusion of membrane transporters, specifically for T4, between the thyroid gland and the bloodstream. They explored two ways to model these transporters:
- Michaelis-Menten kinetics: A non-linear approach that captures the saturation behavior of transporters.
- Linear approximation: A simplified approach assuming transporter activity is directly proportional to T4 concentration, valid when T4 levels are much lower than the transporter's saturation constant.
Unknown model parameters, particularly those related to transporter activity and hormone conversion, were identified through constrained parameter optimization, fitting the model's steady-state hormone concentrations to real patient data. Dynamic simulations were then performed to observe the system's behavior under these conditions.
Datasets
The study utilized real hormone measurement data from:
- Healthy individuals: 81 untreated healthy individuals from Dietrich (2001).
- AHDS patients: Mean measured hormone concentrations from Groeneweg et al. (2020) and Schwartz et al. (2005).
- Rodent models: Data from Mct8 knockout mice studies (e.g., Trajkovic-Arsic et al., 2010a) were used for comparison and validation.
No new datasets were introduced.
Key Insights and Ideas
- T4 Retention as a Primary Driver: The model strongly supports the hypothesis that impaired T4 transport and subsequent retention within thyroid cells is a key, and potentially sufficient, mechanism to explain the characteristic hormonal profile of AHDS.
- Model Simplicity vs. Accuracy: Both the complex Michaelis-Menten and the simpler linear models of membrane transporters yielded similar results, suggesting that the linear approximation might be sufficient for understanding the core dynamics of AHDS in this context, simplifying future modeling efforts.
- Beyond Extrathyroidal Explanations: While previous hypotheses often invoked extrathyroidal mechanisms (like renal contributions) to explain low FT4, the model suggests that T4 retention within the thyroid itself could be the primary cause, even for the low FT4 levels.
Key Results
- Simulated AHDS Phenotype: Dynamic simulations of the model with damaged membrane transporters accurately replicated the observed hormonal imbalances in AHDS patients (higher TSH and FT3, lower FT4) compared to healthy individuals.
- Elevated Intracellular T4: The model showed an approximately 11-fold increase in T4 concentration within thyroid cells (T4,th) for simulated AHDS patients compared to healthy individuals, aligning with experimental observations in mouse models.
- Model Validation: The model's predictions for healthy individuals remained within established reference ranges, indicating the robustness of the extended model.
- Parameter Identification: The study successfully identified key parameters, demonstrating that the model can be calibrated to real-world data. The linear approximation parameter ($k_1$) was found to be consistent with the ratio of parameters ($G_{MCT8}/K_{MCT8}$) from the Michaelis-Menten model.
Limitations and Future Work
- Kidney's Role: The current model aggregates peripheral organs into a single "periphery" component. The authors acknowledge that an explicit representation of the kidney, which has been suggested to play a role in AHDS, would allow for a more precise investigation of extrathyroidal mechanisms.
- Individualization: The model represents a "generic" healthy individual and a "generic" AHDS patient. Future work could focus on individualizing the model to account for the diverse phenotypes observed in AHDS patients.
- Dynamic Measurements: The current parameter identification relies on steady-state data. The authors suggest that using dynamic hormone measurements would provide richer information and potentially improve parameter identification and model individualization.
- Further Phenomena: Investigating other aspects of MCT8 deficiency, such as the effects of combined deficiencies (e.g., Mct8/D1 double knockout mice), could further deepen understanding.
- Parameter Uniqueness: For healthy individuals, parameter identification was not entirely unique, suggesting that dynamic measurements might be crucial for resolving this.
Summary updated